Deep Research (Multi-Agent Research Orchestration)

Splits a research goal into parallel sub-goals, runs them as headless `claude -p` subprocesses, and aggregates plus chapter-by-chapter polishes the results into a standalone report file.

AutomationAdvanced★ 1,658⑂ 241AI score 10/10Last updated: Sep 27, 2026

What it does

Breaks a broad research goal into parallelizable sub-goals (link lists, topic clusters, time slices, regions, dataset shards).
Launches each sub-goal as a headless claude -p subprocess (8 concurrent by default) with scoped --allowedTools, per-task timeouts and per-task logs.
Creates a run directory .research/<name>/ with prompts/, raw/, child_outputs/, logs/ so raw sources and execution traces stay auditable.
Aggregates child outputs into a base draft via script, redesigns the outline, then iterates chapter by chapter into an insight-driven final report written to a file (never pasted in full into chat).
Treats all scraped content as untrusted data and explicitly ignores injected instructions; failed or timed-out children are reported with cause and follow-up suggestions.

Who it's for

Researchers, strategists and consultants who need multi-day-depth market, competitor or industry analysis.
Analysts batch-processing dozens of links or datasets into a synthesized deliverable.
Claude Code CLI power users who want parallel sub-agents plus traceable evidence trails.

Examples

"Deep research the EV charging infrastructure market" → scouting pass with real samples → 12 sub-goals split by operator/region/policy → chapter-polished analytical report.
Hand it 15 competitor pricing/feature URLs: each becomes a sub-task that scrapes and summarizes, ending in a comparison table plus strategic implications.
Slice three years of papers, blogs and release notes for a technology by year, then synthesize a long-form timeline report.

· · · Install guide · · ·

Try it now, no install

Paste this into Claude to use the skill without installing anything.

Read the instructions in this file and follow them to help me:
https://raw.githubusercontent.com/feiskyer/claude-code-settings/HEAD/skills/deep-research/SKILL.md

What I want: (describe your task here)

If Claude can't open the link, open it yourself and paste the contents instead.

↓ If it works for you, download the ZIP below and install it. Then it runs on its own — no pasting each time.

Install in the Claude app (no terminal)

Download the ZIP with the button below.
In Claude, open Settings → Capabilities and turn on 'Code execution and file creation'. (one time)
Go to Customize → Skills → + → 'Upload a skill' and upload the ZIP.

↓ Download ZIP

Install in Claude Code

Let Claude do it — paste this into Claude Code

Install the skill I found on Claude Skill Mart.
Copy the skills/deep-research folder from the GitHub repo feiskyer/claude-code-settings into my ~/.claude/skills/feiskyer-deep-research/.
When it's done, tell me in one line what this skill can do.

Install with a command instead

git clone https://github.com/feiskyer/claude-code-settings.git /tmp/ccs && mkdir -p ~/.claude/skills && cp -r /tmp/ccs/skills/deep-research ~/.claude/skills/

⚠ This is a third-party skill. Check the source repository before installing.

Open a terminal and confirm Claude Code CLI is installed: claude --version (install Claude Code first if missing).
Clone the repo: git clone https://github.com/feiskyer/claude-code-settings.git /tmp/ccs
Create the skills directory: mkdir -p ~/.claude/skills
Copy the skill: cp -r /tmp/ccs/skills/deep-research ~/.claude/skills/
(Recommended) Configure the firecrawl or exa MCP server for better web collection, and install GNU Parallel for large runs (brew install parallel or apt install parallel).
Restart Claude Code, then ask for "deep research on ..." or "wide research" to trigger the skill.
The skill will present scouting samples and an execution plan and wait for approval — reply "go" or "execute" to start the parallel run.

View source on GitHub ↗License: MIT